Tester
Lipp Laboratory, Department of Environmental Health Science, University of Georgia
2020-09-30
Load all required libraries.
library(tidyverse)## Warning: package 'tidyverse' was built under R version 3.6.3## -- Attaching packages ------------------------------------------------------------------ tidyverse 1.3.0 --## v ggplot2 3.3.2 v purrr 0.3.4
## v tibble 3.0.3 v dplyr 1.0.0
## v tidyr 1.1.0 v stringr 1.4.0
## v readr 1.3.1 v forcats 0.5.0## Warning: package 'ggplot2' was built under R version 3.6.3## Warning: package 'tibble' was built under R version 3.6.3## Warning: package 'readr' was built under R version 3.6.3## Warning: package 'dplyr' was built under R version 3.6.3## Warning: package 'forcats' was built under R version 3.6.3## -- Conflicts --------------------------------------------------------------------- tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()library(plotly)## Warning: package 'plotly' was built under R version 3.6.3##
## Attaching package: 'plotly'## The following object is masked from 'package:ggplot2':
##
## last_plot## The following object is masked from 'package:stats':
##
## filter## The following object is masked from 'package:graphics':
##
## layoutlibrary(broom)## Warning: package 'broom' was built under R version 3.6.3Read in raw data from RDS.
raw_data <- readRDS("./n1_n2_cleaned_cases.rds")Make a few small modifications to names and data for visualizations.
final_data <- raw_data %>% mutate(log_copy_per_L = log10(mean_copy_num_L)) %>%
rename(Facility = wrf) %>%
mutate(Facility = recode(Facility,
"NO" = "WRF A",
"MI" = "WRF B",
"CC" = "WRF C"))Seperate the data by gene target to ease layering in the final plot
#make three data layers
only_positives <<- subset(final_data, (!is.na(final_data$Facility)))
only_n1 <- subset(only_positives, target == "N1")
only_n2 <- subset(only_positives, target == "N2")
only_background <<-final_data %>%
select(c(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke)) %>%
group_by(date) %>% summarise_if(is.numeric, mean)
#specify fun colors
background_color <- "#7570B3"
seven_day_ave_color <- "#E6AB02"
marker_colors <- c("N1" = '#1B9E77',"N2" ='#D95F02')#remove facilty C for now
only_n1 <- only_n1[!(only_n1$Facility == "WRF C"),]
only_n2 <- only_n2[!(only_n2$Facility == "WRF C"),]Build the main plot
#first layer is the background epidemic curve
p1 <- only_background %>%
plotly::plot_ly() %>%
plotly::add_trace(x = ~date, y = ~new_cases_clarke,
type = "bar",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Daily Cases: ', new_cases_clarke),
alpha = 0.5,
name = "Daily Reported Cases",
color = background_color,
colors = background_color,
showlegend = FALSE) %>%
layout(yaxis = list(title = "Clarke County Daily Cases", showline=TRUE)) %>%
layout(legend = list(orientation = "h", x = 0.2, y = -0.3))
#renders the main plot layer two as seven day moving average
p1 <- p1 %>% plotly::add_trace(x = ~date, y = ~X7_day_ave_clarke,
type = "scatter",
mode = "lines",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Seven-Day Moving Average: ', X7_day_ave_clarke),
name = "Seven Day Moving Average Athens",
line = list(color = seven_day_ave_color),
showlegend = FALSE)
#renders the main plot layer three as positive target hits
p2 <- plotly::plot_ly() %>%
plotly::add_trace(x = ~date, y = ~mean_copy_num_L,
type = "scatter",
mode = "markers",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Facility: ', Facility,
'</br> Target: ', target,
'</br> Copies/L: ', round(mean_copy_num_L, digits = 2)),
data = only_n1,
symbol = ~Facility,
marker = list(color = '#1B9E77', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_trace(x = ~date, y = ~mean_copy_num_L,
type = "scatter",
mode = "markers",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Facility: ', Facility,
'</br> Target: ', target,
'</br> Copies/L: ', round(mean_copy_num_L, digits = 2)),
data = only_n2,
symbol = ~Facility,
marker = list(color = '#D95F02', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
layout(yaxis = list(title = "SARS CoV-2 Copies/L",
showline = TRUE,
type = "log",
dtick = 1,
automargin = TRUE)) %>%
layout(legend = list(orientation = "h", x = 0.2, y = -0.3))
#adds the limit of detection dashed line
p2 <- p2 %>% plotly::add_segments(x = as.Date("2020-03-14"),
xend = ~max(date + 10),
y = 3571.429, yend = 3571.429,
opacity = 0.35,
line = list(color = "black", dash = "dash")) %>%
layout(annotations = list(x = as.Date("2020-03-28"), y = 3.8, xref = "x", yref = "y",
text = "Limit of Detection", showarrow = FALSE))
p1## Warning: `arrange_()` is deprecated as of dplyr 0.7.0.
## Please use `arrange()` instead.
## See vignette('programming') for more help
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_warnings()` to see where this warning was generated.## Warning: Ignoring 1 observations p2## Warning: `group_by_()` is deprecated as of dplyr 0.7.0.
## Please use `group_by()` instead.
## See vignette('programming') for more help
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_warnings()` to see where this warning was generated.p_combined <-
plotly::subplot(p2,p1, # plots to combine, top to bottom
nrows = 2,
heights = c(.6,.4), # relative heights of the two plots
shareX = TRUE, # plots will share an X axis
titleY = TRUE
) %>%
# create a vertical "spike line" to compare data across 2 plots
plotly::layout(
xaxis = list(
spikethickness = 1,
spikedash = "dot",
spikecolor = "black",
spikemode = "across+marker",
spikesnap = "cursor"
),
yaxis = list(spikethickness = 0)
)## Warning: Ignoring 1 observationsp_combinedSave the plot to pull into the index
save(p_combined, file = "./plotly_fig.rda")Save an htmlwidget for website embedding
htmlwidgets::saveWidget(p_combined, "plotly_fig.html")Build loess smoothing figures figures
#create smoothing data frames
#n1
smooth_n1 <- only_n1 %>% select(-c(Facility)) %>%
group_by(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke) %>%
summarize(sum_copy_num_L = sum(mean_total_copies)) %>%
ungroup() %>%
mutate(log_sum_copies_L = log10(sum_copy_num_L)) %>%
mutate(target = "N1")## `summarise()` regrouping output by 'date', 'cases_cum_clarke', 'new_cases_clarke', 'X7_day_ave_clarke' (override with `.groups` argument)#n2
smooth_n2 <- only_n2 %>% select(-c(Facility)) %>%
group_by(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke) %>%
summarize(sum_copy_num_L = sum(mean_total_copies)) %>%
ungroup() %>%
mutate(log_sum_copies_L = log10(sum_copy_num_L)) %>%
mutate(target = "N2")## `summarise()` regrouping output by 'date', 'cases_cum_clarke', 'new_cases_clarke', 'X7_day_ave_clarke' (override with `.groups` argument)#add trendlines
#extract data from geom_smooth
#n1 extract
# *********************************span 0.6***********************************
#*****************Must always update the n = TOTAL NUMBER OF DAYS*************************
extract_n1 <- ggplot(smooth_n1, aes(x = date, y = log_sum_copies_L)) +
stat_smooth(aes(outfit=fit_n1<<-..y..), method = "loess", color = '#1B9E77',
span = 0.6, n = 106)## Warning: Ignoring unknown aesthetics: outfit#n2 extract
extract_n2 <- ggplot(smooth_n2, aes(x = date, y = log_sum_copies_L)) +
stat_smooth(aes(outfit=fit_n2<<-..y..), method = "loess", color = '#1B9E77',
span = 0.6, n = 106)## Warning: Ignoring unknown aesthetics: outfit#look at the fits to align dates and total observations
#n1
extract_n1## `geom_smooth()` using formula 'y ~ x'## Warning: Removed 1 rows containing non-finite values (stat_smooth).
fit_n1## [1] 10.76551 10.96873 11.16479 11.35366 11.53527 11.70958 11.87653 12.03608
## [9] 12.18812 12.33234 12.46893 12.59816 12.72028 12.83554 12.94421 13.04655
## [17] 13.14370 13.23611 13.32296 13.40343 13.47673 13.54205 13.59856 13.64533
## [25] 13.68134 13.70726 13.72394 13.73223 13.73299 13.72709 13.71537 13.68835
## [33] 13.64002 13.57642 13.50356 13.42751 13.35428 13.28992 13.23557 13.16787
## [41] 13.08774 13.00096 12.91333 12.83063 12.75867 12.70322 12.64411 12.56329
## [49] 12.46977 12.37257 12.28069 12.20315 12.14896 12.12205 12.09857 12.07632
## [57] 12.05792 12.04601 12.04320 12.05211 12.07537 12.12858 12.21811 12.33281
## [65] 12.46151 12.59305 12.71627 12.81998 12.90065 12.99446 13.10223 13.21738
## [73] 13.33331 13.44342 13.54114 13.61985 13.68980 13.76353 13.83799 13.91010
## [81] 13.97679 14.03499 14.08163 14.11557 14.14605 14.17374 14.19743 14.21591
## [89] 14.22799 14.23247 14.22814 14.21668 14.20034 14.17879 14.15168 14.11868
## [97] 14.07945 14.03365 13.98103 13.92196 13.85659 13.78501 13.70728 13.62350
## [105] 13.53375 13.43810#n2
extract_n2## `geom_smooth()` using formula 'y ~ x'## Warning: Removed 1 rows containing non-finite values (stat_smooth).
fit_n2## [1] 10.63701 10.85591 11.06806 11.27345 11.47207 11.66390 11.84894 12.02719
## [9] 12.19862 12.36322 12.52098 12.67192 12.81603 12.95332 13.08379 13.20744
## [17] 13.32604 13.44061 13.55013 13.65353 13.74978 13.83782 13.91662 13.98610
## [25] 14.05099 14.11087 14.16435 14.21003 14.24648 14.27233 14.28616 14.28060
## [33] 14.25306 14.20928 14.15500 14.09597 14.03792 13.98660 13.94230 13.87866
## [41] 13.79742 13.70585 13.61122 13.52079 13.44181 13.38156 13.32294 13.24843
## [49] 13.16460 13.07805 12.99536 12.92312 12.86791 12.83264 12.79994 12.76811
## [57] 12.73895 12.71427 12.69586 12.68553 12.68506 12.69968 12.73068 12.77398
## [65] 12.82550 12.88114 12.93684 12.98851 13.03647 13.10218 13.18364 13.27431
## [73] 13.36768 13.45721 13.53640 13.59870 13.65642 13.72306 13.79394 13.86437
## [81] 13.92968 13.98518 14.02618 14.05078 14.07231 14.09213 14.10898 14.12158
## [89] 14.12868 14.12900 14.12127 14.10753 14.09031 14.06906 14.04322 14.01226
## [97] 13.97563 13.93277 13.88330 13.82780 13.76655 13.69966 13.62726 13.54949
## [105] 13.46647 13.37833#assign fits to a vector
n1_trend <- fit_n1
n2_trend <- fit_n2
#extract y min and max for each
limits_n1 <- ggplot_build(extract_n1)$data## `geom_smooth()` using formula 'y ~ x'## Warning: Removed 1 rows containing non-finite values (stat_smooth).limits_n1 <- as.data.frame(limits_n1)
n1_ymin <- limits_n1$ymin
n1_ymax <- limits_n1$ymax
limits_n2 <- ggplot_build(extract_n2)$data## `geom_smooth()` using formula 'y ~ x'## Warning: Removed 1 rows containing non-finite values (stat_smooth).limits_n2 <- as.data.frame(limits_n2)
n2_ymin <- limits_n2$ymin
n2_ymax <- limits_n2$ymax
#reassign dataframes (just to be safe)
work_n1 <- smooth_n1
work_n2 <- smooth_n2
#fill in missing dates to smooth fits
work_n1 <- work_n1 %>% complete(date = seq(min(date), max(date), by = "1 day"))
date_vec_n1 <- work_n1$date
work_n2 <- work_n2 %>% complete(date = seq(min(date), max(date), by = "1 day"))
date_vec_n2 <- work_n2$date
#create a new smooth dataframe to layer
smooth_frame_n1 <- data.frame(date_vec_n1, n1_trend, n1_ymin, n1_ymax)
smooth_frame_n2 <- data.frame(date_vec_n2, n2_trend, n2_ymin, n2_ymax)#make plotlys
#plot smooth frames
p3 <- plotly::plot_ly() %>%
plotly::add_lines(x = ~date_vec_n1, y = ~n1_trend,
data = smooth_frame_n1,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n1,
'</br> Median Log Copies: ', round(n1_trend, digits = 2),
'</br> Target: N1'),
line = list(color = '#1B9E77', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_lines(x = ~date_vec_n2, y = ~n2_trend,
data = smooth_frame_n2,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n2,
'</br> Median Log Copies: ', round(n2_trend, digits = 2),
'</br> Target: N2'),
line = list(color = '#D95F02', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_ribbons(x ~date_vec_n1, ymin = ~n1_ymin, ymax = ~n1_ymax,
showlegend = FALSE,
opacity = 0.25,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n1, #leaving in case we want to change
'</br> Max Log Copies: ', round(n1_ymax, digits = 2),
'</br> Min Log Copies: ', round(n1_ymin, digits = 2),
'</br> Target: N1'),
name = "",
line = list(color = '#1B9E77')) %>%
plotly::add_ribbons(x ~date_vec_n2, ymin = ~n2_ymin, ymax = ~n2_ymax,
showlegend = FALSE,
opacity = 0.25,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n2, #leaving in case we want to change
'</br> Max Log Copies: ', round(n2_ymax, digits = 2),
'</br> Min Log Copies: ', round(n2_ymin, digits = 2),
'</br> Target: N2'),
name = "",
line = list(color = '#D95F02')) %>%
layout(yaxis = list(title = "Total Log SARS CoV-2 Copies",
showline = TRUE,
automargin = TRUE)) %>%
layout(xaxis = list(title = "Date")) %>%
plotly::add_segments(x = as.Date("2020-06-24"),
xend = as.Date("2020-06-24"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "Bars Repoen",
hoverinfo = "text",
text = "</br> Bars Reopen",
"</br> 2020-06-24",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-07-09"),
xend = as.Date("2020-07-09"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "Mask Mandate",
hoverinfo = "text",
text = "</br> Mask Mandate",
"</br> 2020-07-09",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-08-20"),
xend = as.Date("2020-08-20"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "</br> Classes Begin",
"</br> 2020-08-20",
hoverinfo = "text",
text = "Classes Begin",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_markers(x = ~date, y = ~log_sum_copies_L,
data = smooth_n1,
hoverinfo = "text",
showlegend = FALSE,
text = ~paste('</br> Date: ', date,
'</br> Actual Log Copies: ', round(log_sum_copies_L, digits = 2)),
marker = list(color = '#1B9E77', size = 6, opacity = 0.65)) %>%
plotly::add_markers(x = ~date, y = ~log_sum_copies_L,
data = smooth_n2,
hoverinfo = "text",
showlegend = FALSE,
text = ~paste('</br> Date: ', date,
'</br> Actual Log Copies: ', round(log_sum_copies_L, digits = 2)),
marker = list(color = '#D95F02', size = 6, opacity = 0.65))
p3## Warning: Ignoring 1 observations
## Warning: Ignoring 1 observationsCreate final trend plot by stacking with epidemic curve
smooth_extract <-
plotly::subplot(p3,p1, # plots to combine, top to bottom
nrows = 2,
heights = c(.6,.4), # relative heights of the two plots
shareX = TRUE, # plots will share an X axis
titleY = TRUE
) %>%
# create a vertical "spike line" to compare data across 2 plots
plotly::layout(
xaxis = list(
spikethickness = 1,
spikedash = "dot",
spikecolor = "black",
spikemode = "across+marker",
spikesnap = "cursor"
),
yaxis = list(spikethickness = 0)
)## Warning: Ignoring 1 observations
## Warning: Ignoring 1 observations
## Warning: Ignoring 1 observationssmooth_extractsave(smooth_extract, file = "./smooth_extract.rda")